The breast is an estrogen responsive tissue, and most breast tumors appear to be, at least initially, estrogen dependent. Furthermore, the estrogen receptor (ER) is overexpressed not only in breast tumors but even in early premalignant breast lesions as compared with normal breast epithelium, so that it may be associated with progression toward cancer. We have also recently found in several early lesions a unique ER mutant which is hypersensitive to estrogen. In many breast tumors we have found enhanced expression of ER splice variants, particularly an exon 5 deletion variant which is active even without estrogen and confers tamoxifen resistance. Thus it seems likely that derangements in the expression level or structure of ER may well play roles both in the development of breast cancer and in the failure of hormone responsiveness during treatment. We now propose to investigate how ER overexpression arises during early progression of breast lesions, how the appearance of certain variants may contribute to the biological behavior of primary breast tumors, particularly tamoxifen resistance, and whether further ER alterations may be associated with the development of metastases. Our Aims are: 91) To determine the molecular basis for elevated ER expression in hyperplastic breast lesion and breast tumors as compared to adjacent normal breast epithelium, by analysis of the promoter region of the ER and by analysis of the cell-specific ER transactivating factors present which are responsible for the inappropriate activation of ER expression in these lesions. (2) To discover whether expression of the truncated exon 5 ER deletion variant in primary breast cancer is associated with clinical tamoxifen resistance, using tamoxifen-treated patients in the San Antonio Tumor Bank and in a large randomized Swedish adjuvant trial. (3) To study ER alterations in metastatic breast cancers which might be associated with tumor progression, by both direct DNA sequencing and SSCP analyses of paired primary and metastatic tumor specimens. Identified ER alterations will be characterized using our series of unique estrogen-inducible ERE reporters. Cell migration, invasion, and metastatic behavior will also be determined in stable breast cancer cell line transfectants to correlate these features with expression of specific ER alterations. We will also correlate levels of individual ER variants isolated from metastatic breast lesions with clinical outcome. We have already made important discoveries on the occurrence of variant forms of the estrogen receptor and regulation of the receptor in clinical breast cancer. The proposed work should now shed light on the role of Er in early breast lesions. It should also indicate whether the appearance of higher levels of the truncated exon 5 deletion ER splice variant in breast cancer is related to the clinically ominous development of tamoxifen resistance, and suggest which ER variants or mutations presage metastasis.